使用Lucene.NET實(shí)現(xiàn)簡單的站內(nèi)搜索

使用Lucene.NET實(shí)現(xiàn)簡單的站內(nèi)搜索

1. 導(dǎo)入Lucene.NET 開發(fā)包

   Lucene 是apache軟件基金會(huì)一個(gè)開放源代碼的全文檢索引擎工具包，是一個(gè)全文檢索引擎的架構(gòu)，提供了完整的查詢引擎和索引引擎，部分文本分析引擎。Lucene的目的是為軟件開發(fā)人員提供一個(gè)簡單易用的工具包，以方便的在目標(biāo)系統(tǒng)中實(shí)現(xiàn)全文檢索的功能，或者是以此為基礎(chǔ)建立起完整的全文檢索引擎。Lucene.Net 是 .NET 版的Lucene。

   你可以在這里下載到最新的Lucene.NET

2. 創(chuàng)建索引、更新索引、刪除索引

   

3. 搜索，根據(jù)索引查找

   

   

     1 using System;  2 using Lucene.Net.Store;  3 using Lucene.Net.Index;  4 using Lucene.Net.Analysis.PanGu;  5 using Lucene.Net.Documents;  6   7 namespace BLL  8 {  9     class IndexHelper 10     { 11         /// <summary> 12         /// 日志小助手 13         /// </summary> 14         static Common.LogHelper logger = new Common.LogHelper(typeof(SearchBLL)); 15         /// <summary> 16         /// 索引保存的位置，保存在配置文件中從配置文件讀取 17         /// </summary> 18         static string indexPath = Common.ConfigurationHelper.AppSettingMapPath("IndexPath"); 19          20         /// <summary> 21         /// 創(chuàng)建索引文件或更新索引文件 22         /// </summary> 23         /// <param name="item">索引信息</param> 24         public static void CreateIndex(Model.HelperModel.IndexFileHelper item) 25         { 26             try 27             { 28                 //索引存儲(chǔ)庫 29                 FSDirectory directory = FSDirectory.Open(new System.IO.DirectoryInfo(indexPath), new NativeFSLockFactory()); 30                 //判斷索引是否存在 31                 bool isUpdate = IndexReader.IndexExists(directory); 32                 if (isUpdate) 33                 { 34                     //如果索引目錄被鎖定（比如索引過程中程序異常退出），則首先解鎖 35                     if (IndexWriter.IsLocked(directory)) 36                     { 37                         //解鎖索引庫 38                         IndexWriter.Unlock(directory); 39                     } 40                 } 41                 //創(chuàng)建IndexWriter對象，添加索引 42                 IndexWriter writer = new IndexWriter(directory, new PanGuAnalyzer(), !isUpdate, Lucene.Net.Index.IndexWriter.MaxFieldLength.UNLIMITED); 43                 //獲取新聞 title部分 44                 string title = item.FileTitle; 45                 //獲取新聞主內(nèi)容 46                 string body = item.FileContent; 47                 //為避免重復(fù)索引，所以先刪除number=i的記錄，再重新添加 48                 //尤其是更新的話，更是必須要先刪除之前的索引 49                 writer.DeleteDocuments(new Term("id", item.FileName)); 50                 //創(chuàng)建索引文件 Document 51                 Document document = new Document(); 52                 //只有對需要全文檢索的字段才ANALYZED 53                 //添加id字段 54                 document.Add(new Field("id", item.FileName, Field.Store.YES, Field.Index.NOT_ANALYZED)); 55                 //添加title字段 56                 document.Add(new Field("title", title, Field.Store.YES, Field.Index.NOT_ANALYZED)); 57                 //添加body字段 58                 document.Add(new Field("body", body, Field.Store.YES, Field.Index.ANALYZED, Lucene.Net.Documents.Field.TermVector.WITH_POSITIONS_OFFSETS)); 59                 //添加url字段 60                 document.Add(new Field("url", item.FilePath, Field.Store.YES, Field.Index.NOT_ANALYZED)); 61                 //寫入索引庫 62                 writer.AddDocument(document); 63                 //關(guān)閉資源 64                 writer.Close(); 65                 //不要忘了Close，否則索引結(jié)果搜不到 66                 directory.Close(); 67                 //記錄日志 68                 logger.Debug(String.Format("索引{0}創(chuàng)建成功",item.FileName)); 69             } 70             catch (SystemException ex) 71             { 72                 //記錄錯(cuò)誤日志 73                 logger.Error(ex); 74                 throw; 75             } 76             catch (Exception ex) 77             { 78                 //記錄錯(cuò)誤日志 79                 logger.Error(ex); 80                 throw; 81             } 82         } 83  84         /// <summary> 85         /// 根據(jù)id刪除相應(yīng)索引 86         /// </summary> 87         /// <param name="guid">要?jiǎng)h除的索引id</param> 88         public static void DeleteIndex(string guid) 89         { 90             try 91             { 92                 ////索引存儲(chǔ)庫 93                 FSDirectory directory = FSDirectory.Open(new System.IO.DirectoryInfo(indexPath), new NativeFSLockFactory()); 94                 //判斷索引庫是否存在索引 95                 bool isUpdate = IndexReader.IndexExists(directory); 96                 if (isUpdate) 97                 { 98                     //如果索引目錄被鎖定（比如索引過程中程序異常退出），則首先解鎖 99                     if (IndexWriter.IsLocked(directory))100                     {101                         IndexWriter.Unlock(directory);102                     }103                 }104                 IndexWriter writer = new IndexWriter(directory, new PanGuAnalyzer(), !isUpdate, Lucene.Net.Index.IndexWriter.MaxFieldLength.UNLIMITED);105                 //刪除索引文件106                 writer.DeleteDocuments(new Term("id", guid));107                 writer.Close();108                 directory.Close();//不要忘了Close，否則索引結(jié)果搜不到109                 logger.Debug(String.Format("刪除索引{0}成功", guid));110             }111             catch (Exception ex)112             {113                 //記錄日志114                 logger.Error(ex);115                 //拋出異常116                 throw;117             }118         }119     }120 }

   IndexHelper 添加、更新、刪除索引

     1 using Lucene.Net.Analysis;  2 using Lucene.Net.Analysis.PanGu;  3 using Lucene.Net.Documents;  4 using Lucene.Net.Index;  5 using Lucene.Net.Search;  6 using Lucene.Net.Store;  7 using Model.HelperModel;  8 using System;  9 using System.Collections.Generic; 10  11 namespace BLL 12 { 13     public static class SearchBLL 14     { 15         //一個(gè)類中可能會(huì)有多處輸出到日志，多處需要記錄日志，常將logger做成static 靜態(tài)變量 16         /// <summary> 17         /// 日志助手 18         /// </summary> 19         static Common.LogHelper logger = new Common.LogHelper(typeof(SearchBLL)); 20         /// <summary> 21         /// 索引保存位置 22         /// </summary> 23         static string indexPath = Common.ConfigurationHelper.AppSettingMapPath("IndexPath"); 24         /// <summary> 25         /// 搜索 26         /// </summary> 27         /// <param name="keyWords">用戶搜索的關(guān)鍵詞</param> 28         /// <returns>返回搜索的結(jié)果</returns> 29         public static List<SearchResult> Search(string keywords) 30         { 31             try 32             { 33                 //索引存儲(chǔ)庫 34                 FSDirectory directory = FSDirectory.Open(new System.IO.DirectoryInfo(indexPath), new NoLockFactory()); 35                 //創(chuàng)建IndexReader對象 36                 IndexReader reader = IndexReader.Open(directory, true); 37                 //創(chuàng)建IndexSearcher對象 38                 IndexSearcher searcher = new IndexSearcher(reader); 39                 //新建PhraseQuery 查詢對象 40                 PhraseQuery query = new PhraseQuery(); 41                 //把用戶輸入的關(guān)鍵詞進(jìn)行拆詞 42                 foreach (string word in SplitWord(keywords)) 43                 { 44                     //添加搜索關(guān)鍵詞 45                     query.Add(new Term("body", word)); 46                 } 47                 //設(shè)置分詞間距為100字之內(nèi) 48                 query.SetSlop(100); 49                 TopScoreDocCollector collector = TopScoreDocCollector.create(1000, true); 50